def main():
csv = pd.read_csv('city_aqi.csv')
print(csv[['City', 'AQI']].sort_values('AQI').head(4))
print(csv.info())
print(csv.head())
print('AQI最大值：', csv['AQI'].max())
print('AQI最小值：', csv['AQI'].min())
print('AQI均值：', csv['AQI'].mean())
top5_cities = csv.sort_values(by=['AQI']).head(5)
print('空气质量最好的5个城市：')
print(top5_cities)
top5_bottom_cities = csv.sort_values(by=['AQI'], ascending=False).head(5)
print('空气质量最差的5个城市：')
print(top5_bottom_cities)
print(csv[csv['AQI'] > 40])
top5_cities.plot(kind='bar',
x='City',
y='AQI',
title='空气质量最好的5个城市',
figsize=(10, 10))
plt.savefig('top5_aqi.png')
plt.show()
def __init__(self, file, mono=True, cap_train=None, sort=False):
if sort:
csv = pd.read_csv(file)
empty_counts = csv['puzzle'].apply(lambda p: p.count('0'))
csv['num_empty'] = empty_counts
csv.sort_values(by="num_empty", inplace=True)
if cap_train is not None:
csv = csv.head(cap_train)
self.data = csv
else:
self.data = pd.read_csv(file, nrows=cap_train)
self.mono = mono
self.cap_train = cap_train
self.edges = sudoku_edges()
def read_tasks(self, mapper: DataMapper):
self.data_mapper = mapper
csv = pd.read_csv(self.filename, delimiter="\t")
# 1 means to operate on the rows, not the columns.
csv["Time Remaining"] = csv.apply(self.data_mapper.map_submission,
axis=1)
csv["Adjusted Priority"] = csv.apply(
self.data_mapper.map_adjusted_priority, axis=1)
csv = csv.sort_values(by=["Adjusted Priority"])
csv["Sprint"] = csv["Estimate"].apply(self.data_mapper.map_sprints)
self.data_mapper.sprint_manager.reset()
csv["Confidence"] = csv.apply(self.data_mapper.map_sprint_confidence,
axis=1)
csv["On track"] = csv["Confidence"].apply(
self.data_mapper.map_on_track())
csv["Submitted"] = csv.apply(self.data_mapper.map_human_time_submitted,
axis=1)
csv["Due"] = csv.apply(self.data_mapper.map_human_time_due, axis=1)
return csv
# In[21]:
def infra_density(num_o_houses, mask):
area = (np.count_nonzero(mask)) * 10
den = num_o_houses / area
return den
# In[ ]:
# In[11]:
csv = pd.read_csv(csv_path)
csv.head(10)
x = csv.sort_values(by=['Village Name', 'Census 2011 ID'])
x.head(10)
# In[12]:
status = x.drop([23528], axis=0)
status.head(10)
status['Electrified'][0]
# In[74]:
index = 0
village_name = list()
ndvi = list()
evi_log = list()
def sortSeg(id):
print("sorting %d" % id)
csv = pd.read_csv('../result/segg/seg%d.csv' % id)
csv.sort_values('times', inplace=True, ascending=False)
csv.to_csv("../result/seg/%d.csv" % id, index=False)
"slug":
current["slug"],
"name":
current["name"],
"creator_pseudo":
current["creator"]["pseudo"],
"categories":
"|".join([e["name"] for e in current["categories"]]),
"youtube_url":
extract_url(current["links"], "youtube"),
"twitter_url":
extract_url(current["links"], "twitter"),
"tip_amount":
int(current["parameters"]["tipperAmount"]),
"tip_number":
int(current["parameters"]["tipperNumber"]),
})
with open(FILENAME, "a") as f:
writer = csv.DictWriter(f, data[0].keys(), lineterminator="\n")
if f.tell() == 0:
writer.writeheader()
writer.writerows(data)
csv = pd.read_csv(FILENAME, parse_dates=["date"])
csv.drop_duplicates(subset=["date", "slug"], keep="last", inplace=True)
csv.sort_values(by=["date", "slug"], inplace=True)
csv.to_csv(FILENAME, index=False)
def clustering(file_name,times_thr, path_thr,n_clusters) :
"""
This function return a dictionary, which contains a cluster (as a key) and
the associated plates (as values of that key)
- **parameters**, **types**, **return** and **return types**::
:param file_name: file name
:param times_thr: threshold of times, it is going to select greater values
:param path_thr: threshold of the lenght of the paths, it is going to select greater values
:param n_cluster: number of clusters for Kmeans algorithm
:type file_name: string
:type times_thr: int
:type path_thr: int
:type n_cluster: int
:return: return a dictionary, which contains a cluster (as a key) and the associated plates (as values of that key)
:rtype: dictionary (int, [int,..])
"""
csv= pd.read_csv(file_name, sep=',',index_col=None)
csv = csv.loc[csv['volte']>=times_thr]
sorted_csv = csv.sort_values('targa')
csv_np = sorted_csv.values
plates = sorted_csv['targa'].unique()
lista = list()
total_paths = np.array([])
for count,plate in enumerate(plates) :
paths =csv_np[np.where(csv_np[:,0] == plate),1][0]
volte =csv_np[np.where(csv_np[:,0] == plate),2][0]
lista.insert(count, [plate])
i=0
while i < len(paths) :
if len(paths[i].split('-')) <= path_thr :
i+=1
continue
lista[count].append(paths[i])
total_paths= np.append(total_paths,paths[i])
lista[count].append(volte[i])
i+=1
total_paths = np.unique(total_paths)
plates = np.array([],dtype=int)
lista = [x for x in lista if len(x)>1]
for x in lista :
plates = np.append(plates,x[0])
data = np.zeros(shape=(len(lista),len(total_paths)), dtype = np.int8)
print "selected plates: "+str(plates)+"\nselected paths: " + str(total_paths)+"\nmatrix shape: " + str(data.shape)
for count,element in enumerate(lista) :
i=1
while i < len(element) :
index = total_paths.tolist().index(element[i])
data[count,index] = element[i+1]
i+=2
print data
clusterer= KMeans(n_clusters=n_clusters, random_state=10)
kmeans = clusterer.fit(data)
cluster_labels = clusterer.fit_predict(data)
silhouette_avg = silhouette_score(data, cluster_labels)
print "For n_clusters =", n_clusters,"The average silhouette_score is :", silhouette_avg
sample_silhouette_values = silhouette_samples(data, cluster_labels)
clusters_map = {}
for cluster in kmeans.labels_ :
clusters_map[cluster] = []
for i,plate in enumerate(lista) :
clusters_map[kmeans.labels_[i]].append(plate[0])
print "\ncluster\tplates"
for k,v in clusters_map.items():
print k,"\t",v
fig, (ax1,ax2) = plt.subplots(1, 2)
fig.set_size_inches(18, 7)
ax2.set_title("The visualization of the clustered data.")
ax2.set_xlabel("Feature space for the 1st feature")
ax2.set_ylabel("Feature space for the 2nd feature")
colors = cm.spectral(cluster_labels.astype(float) / n_clusters)
ax2.scatter(data[:, 0], data[:, 1], marker='.', s=30, lw=0, alpha=0.7,c=colors, edgecolor='k')
centers = clusterer.cluster_centers_
ax2.scatter(centers[:, 0], centers[:, 1], marker='o',c="white", alpha=1, s=200, edgecolor='k')
for i, c in enumerate(centers):
ax2.scatter(c[0], c[1], marker='$%d$' % i, alpha=1, s=50, edgecolor='k')
plt.suptitle(("KMeans clustering on sample data "
"with n_clusters = %d\nThe average silhouette_score is %0.4f" % (n_clusters,silhouette_avg)),
fontsize=14, fontweight='bold')
ax1.bar(clusters_map.keys(),[len(clusters_map[x]) for x in clusters_map.keys() ],color='r')
ax1.set_ylabel("number of cars")
ax1.set_xlabel("clusters")
ax1.set_title("K-Means clustering")
plt.show()